huggingface / trl

Train transformer language models with reinforcement learning.
http://hf.co/docs/trl
Apache License 2.0
10.13k stars 1.28k forks source link

📃 Fix description for parameter "generate_during_eval" in dpo_config #2364

Closed dakru012 closed 4 days ago

dakru012 commented 4 days ago

What does this PR do?

The description of the "generate_during_eval" parameter is incorrectly that of "truncation_mode" in the DPO config. I aligned it with other similar trainer configs such as BCO or KTO.

Before submitting

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR.

qgallouedec commented 4 days ago

Thanks!

HuggingFaceDocBuilderDev commented 4 days ago

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.